1. No category

DEPARTAMENTO DE ECONOMÍA
PONTIFICIA UNIVERSIDAD CATÓLICA DEL PERÚ
DEPARTAMENTO DE ECONOMÍA
PONTIFICIA UNIVERSIDAD CATÓLICA DEL PERÚ
DOCUMENTO DE TRABAJO N° 389
DEPARTAMENTO DE ECONOMÍA
DRIVING ECONOMIC FLUCTUATIONS IN PERU:
PONTIFICIA
UNIVERSIDAD CATÓLICA:DE?L PERÚ
THE ROLE OF THE TERMS
OF TRADE
DEPARTAMENTO
DE ECONOMÍA
Gabriel Rodríguez y Pierina
Villanueva
PONTIFICIA UNIVERSIDAD CATÓLICA DEL PERÚ
DEPARTAMENTO DE ECONOMÍA
PONTIFICIA UNIVERSIDAD CATÓLICA DEL PERÚ
DEPARTAMENTO DE ECONOMÍA
PONTIFICIA UNIVERSIDAD CATÓLICA DEL PERÚ
DEPARTAMENTO DE ECONOMÍA
PONTIFICIA UNIVERSIDAD CATÓLICA DEL PERÚ
DEPARTAMENTO DE ECONOMÍA
PONTIFICIA UNIVERSIDAD CATÓLICA DEL PERÚ
DEPARTAMENTO DE ECONOMÍA
PONTIFICIA UNIVERSIDAD CATÓLICA DEL PERÚ
DEPARTAMENTO DE ECONOMÍA
PONTIFICIA UNIVERSIDAD CATÓLICA DEL PERÚ
DEPARTAMENTO DE ECONOMÍA
DEPARTAMENTO
DE ECONOMÍA
DOCUMENTO DE TRABAJO N° 389
DRIVING ECONOMIC FLUCTUATIONS IN PERU:
THE ROLE OF THE TERMS OF TRADE
Gabriel Rodríguez y Pierina Villanueva
Diciembre, 2014
DEPARTAMENTO
DE ECONOMÍA
DOCUMENTO DE TRABAJO 389
http://files.pucp.edu.pe/departamento/economia/DDD389.pdf
© Departamento de Economía – Pontificia Universidad Católica del Perú,
© Gabriel Rodríguez y Pierina Villanueva
Av. Universitaria 1801, Lima 32 – Perú.
Teléfono: (51-1) 626-2000 anexos 4950 - 4951
Fax: (51-1) 626-2874
[email protected]
www.pucp.edu.pe/departamento/economia/
Encargado de la Serie: Jorge Rojas Rojas
Departamento de Economía – Pontificia Universidad Católica del Perú,
[email protected]
Gabriel Rodríguez y Pierina Villanueva
Driving Economic Fluctuations in Peru: The Role of the Terms of Trade
Lima, Departamento de Economía, 2014
(Documento de Trabajo 389)
PALABRAS CLAVE: Fluctuaciones, Producto, Consumo, Inversión Privada,
Inversión Pública, Términos de Intercambio, Choques Permanentes y
Transitorios, Choques Domésticos y Foráneos, Pequeña Economía Abierta
en Desarrollo, tendencias Comunes, Cointegración.
Las opiniones y recomendaciones vertidas en estos documentos son responsabilidad de sus
autores y no representan necesariamente los puntos de vista del Departamento Economía.
Hecho el Depósito Legal en la Biblioteca Nacional del Perú Nº 2015-02358.
ISSN 2079-8466 (Impresa)
ISSN 2079-8474 (En línea)
Impreso en Kolores Industria Gráfica E.I.R.L.
Jr. La Chasca 119, Int. 264, Lima 36, Perú.
Tiraje: 100 ejemplares
Driving Economic Fluctuations in Peru: The Role of the Terms of
Trade
Gabriel Rodríguez
Ponti…cia Universidad Católica del Perú
Pierina Villanueva Vega
Ponti…cia Universidad Católica del Perú
Abstract
This paper has four objectives. Firstly, to verify the existence of long-term relationships between
the groups of variables analyzed (product, consumption, private investment, public investment, and
terms of trade). Secondly, to analyze the role of public and private investment, as well as the role of
the terms of trade in the Peruvian economy’s economic ‡uctuations. Thirdly, to identify domestic
and foreign shocks, as well as the degree of importance of both in the economy’s ‡uctuations.
Finally, to identify the role and the impact of permanent and transitory shocks in the economic
‡uctuations of an emerging economy such as Peru. To achieve these objectives, we follow the
focuses of King et al. (1991), Mellander et al. (1992), and Warne (1993); additionally, the analysis
disaggregates the total public and private investment. The primary result is that the permanent
shocks of the terms of trade (foreign shocks) account for most of the ‡uctuations in product,
consumption, private investment, and public investment. This result appears more pronounced
as the time horizon approaches the long-term. The transitory shocks, for their part, explain the
‡uctuations of some variables only in the short term.
JEL Classi…cation: E32, F41, F43, C32.
Keywords: Fluctuations, Output, Consumption, Private Investment, Public Investment, Terms of
Trade, Permanent and Transitory Shocks, Domestic and Foreign Shocks, Developing Small Open
Economy, Common Trends, Cointegration.
Resumen
Este trabajo tiene cuatro objetivos. En primer lugar, veri…car la existencia de relaciones de
largo plazo entre el grupo de variables analizadas (producto, consumo, inversión privada, inversión pública, y términos de intercambio). En segundo lugar, analizar el rol de la inversión pública
y privada, así como el papel de los términos de intercambio en las ‡uctuaciones de la economía Peruana. En tercer lugar, identi…car los choques domésticos y externos, así como su respectivo grado
de importancia en las ‡uctuaciones de la economía. Por último, identi…car el papel y el impacto de
los choques permanentes y transitorios en las ‡uctuaciones económicas de una economía emergente
como el Perú. Para lograr los objetivos, seguimos los enfoques de King et al. (1991), Mellander et
al. (1992), and Warne (1993); además, el análisis desagrega la inversión pública y privada total. El
resultado principal es que los choques permanentes de los términos de intercambio (choques foráneos) explican la mayor parte de las ‡uctuaciones en el producto, el consumo, la inversión privada
y la inversión pública. Este resultado es más pronunciado a medida que se acerca el horizonte de
tiempo al largo plazo. Los choques transitorios, por su parte, explican las ‡uctuaciones de algunas
variables sólo en el corto plazo.
Classi…cación JEL: E32, F41, F43, C32.
Palabras Claves: Fluctuaciones, Producto, Consumo, Inversión Privada, Inversión Pública, Términos de INtercambio, Choques Permanentes y Transitorios, Choques Domésticos y Foráneos,
Pequeña Economía Abierta en Desarrollo, tendnecias Comunes, Cointegración.
Driving Economic Fluctuations in Peru: The Role of the Terms of
Trade1
Gabriel Rodríguez2
Ponti…cia Universidad Católica del Perú
1
Pierina Villanueva Vega
Ponti…cia Universidad Católica del Perú
Introduction
The recent experience of the Peruvian economy, marked by solid terms of trade, raw material
price increases, and accelerated growth from mid-2000, suggests that terms of trade shocks are
instrumental in explaining the ‡uctuations of emerging economies. The Gross Domestic Product
(GDP) posted an average growth of 6.3% over the period 2002-2007, while the terms of trade
enjoyed six years of sustained growth over the same period. In real terms, from 2000 until just
before the international crisis, the terms of trade rose by 75%. The volume of exports, consumption,
and the product grew at high rates: 7.5% in developing economies compared with 3.1% in advanced
economies; see IMF (2008).
In theory, terms of trade shocks can explain both the cyclical and long term behavior of
economies, depending on the duration of the shocks and the transmission mechanisms. According
to Obstfeldt (1982) and Svensson and Razin (1983), the terms of trade a¤ect purchasing power,
which has impacts on the product and in consequence on consumption, which can be strongly
a¤ected insofar as the shock is more permanent. Another transmission mechanism is the fact that
the gains in the terms of trade induce increased investment, especially in export-oriented industries.
This is all the more evident when the export sector is related to raw materials, which is true for
emerging economies such as Peru.
The relationship between terms of trade and economic ‡uctuations has been studied by authors
such as Mendoza (1995), Easterly et al. (2001), and Becker and Mauro (2006). The dependence
of developing economies on imported capital goods, and the high concentration of raw materials in
the composition of their exports, render the terms of trade relevant to their ‡uctuations. In this
regard, Blattamn et al. (2004) review the relationship between the terms of trade and the behavior
of developing economies. Mendoza (1995) …nds that prominent shocks to the terms of trade are
liable to be long, persistent, and pro-cyclical, and that developing countries are more variable in
terms of their macroeconomic aggregates.
Dancourt et al. (1997) evaluate the signi…cance of external shocks in Peruvian macroeconomic
history over the period 1950-1996. To this end, they construct an external shock indicator that is
directly correlated to the development of the terms of trade, with the exception of the 1990s given
the importance of capital ‡ows to the period. The authors …nd that all of the economic recessions
studied, except for that which began in 1987, coincide with adverse external shocks. Moreover,
1
This paper is drawn from the Thesis of Pierina Villanueva (2013) at the Department of Economics, Ponti…cia
Universidad Católica del Perú. We thank useful comments of Paul Castillo (Central Reserve Bank of Peru). Helpful
comments and conversations with Oscar Dancourt (PUCP) are especially appreciated.
2
Address for Correspondence: Gabriel Rodríguez, Department of Economics, Ponti…cia Universidad Católica del
Perú, Av. Universitaria 1801, Lima 32, Lima, Perú, Telephone: +511-626-2000 (4998), Fax: +511-626-2874. E-Mail
Address: [email protected].
1
they …nd that the …rst half of the 1990s was marked by the biggest and longest-lasting favorable
external shock experienced by the Peruvian economy in the 47 years of analysis. In this regard, the
authors conclude that the role of external shocks cannot be overlooked when attempting to explain
the long term behavior of the Peruvian economy.
Mellander et al. (1992) analyze macroeconomic ‡uctuations in a small and open economy such
as Sweden. To this end, they make use of a common stochastic trends model, in which the terms of
trade, consumption, gross domestic product, and aggregate investment are included. The authors
…nd two cointegration relationships and propose two stochastic trends (domestic and foreign). The
results show that both trends positively in‡uence each of the series analyzed in the long term.
On the other hand, the transitory shocks are signi…cant for the case of the terms of change and
aggregate investment in the short and medium term. Nonetheless, in the long term, both variables
are explained by permanent foreign shocks. In the case of the product, transitory shocks are not
signi…cant, even in the short term. In the long term, the product is explained by permanent
internal and external shocks of a similar magnitude. Moreover, the authors …nd that permanent
domestic shocks have greater in‡uence on consumption than on the long-term product; foreign
shock in‡uences consumption variability almost in its entirety. The results show that permanent
shocks have an important role in economic ‡uctuations in the case of a small and open economy.
At this respect, Kose and Riezman (1999) …nd that 45% of economic ‡uctuations are explained by
shocks arising from international trade in a sample of African countries. Moreover, they …nd that
these shocks explain around 90% of the aggregate investment variation and 80% of the consumption
‡uctuations; see also Kose (2002).
Misas et al. (2003), perform a study on the ‡uctuations of Colombia’s main macroeconomic
variables over the period 1950-2002. The authors employ a common stochastic trends model and
two cointegration relationships, and …nd that the permanent shocks are important in explaining
macroeconomic ‡uctuations, even though they do not display a variance decomposition.
Castillo and Salas (2010), perform an analysis on the Peruvian economy’s economic ‡uctuations
for the quarterly period 1992-2007. The variables studied are the terms of trade, gross domestic
product, consumption, and aggregate investment. The authors estimate a stochastic trends model
to identify the role of economic shocks on macroeconomic variables. Previously, they …nd two
cointegration relationships and pose the existence of two common trends -one foreign and the other
domestic. The results show that both trends have a positive e¤ect on each of the variables for the
long-term. The authors show that the permanent shocks signi…cantly in‡uence the economy, and the
result increases in pertinence in the medium- and long-term. Moreover, they …nd that permanent
foreign shocks account for the greatest proportion of ‡uctuations in product, consumption, and
total investment in the Peruvian economy.
Other studies that …nd signi…cant in‡uence of external shocks on the ‡uctuations of a small,
open economy include Canova (2005), Aguiar and Gopinath (2007), Justiniano and Preston (2008),
and Izquierdo et al. (2008). See also recent works in IMF (2014a, 2014b). Studies that …nd
the reverse include Ahmed and Murphy (1994), Ho¤maister and Roldós (1997), Ho¤maister et al.
(1998), and Lubik and Teo (2005).
This paper has four objectives. Firstly, to verify the existence of long-term relationships between
the group of variables analyzed (GDP, consumption, private investment, public investment, and
terms of trade). Secondly, to analyze the role of public and private investment, as well as the role of
the terms of trade in the Peruvian economy’s economic ‡uctuations. Thirdly, to identify domestic
and foreign shocks, as well as the degree of importance of both in the economy’s ‡uctuations.
2
Finally, to identify the role and the impact of permanent and transitory shocks in the economic
‡uctuations of an emerging economy such as Peru’s. To achieve the objectives, we follow the focuses
of King et al. (1991), Mellander et al. (1992), and Warne (1993), Misas et al. (2003) and Castillo
and Salas (2010); additionally, the analysis disaggregates the total public and private investment.
The primary result is that the permanent shocks of the terms of trade (foreign shock) account for
most of the ‡uctuations in product, consumption, private investment, and public investment. This
result appears more pronounced as the time horizon approaches the long-term. The transitory
shocks, for their part, explain the ‡uctuations of some variables only in the short term.
The paper is structured as follows. Section 2 presents Johansen’s cointegration methodology
(1988, 1991, 1995) and the common stochastic trends of Stock and Watson (1988), Mellander et al.
(1992), and Warne (1993). Section 3 presents and discusses the empirical results of both approaches.
Section 4 presents the main conclusions.
2
Methodology
This study makes use of the cointegration methodology put forward by Johansen (1988, 1995) and
the stochastic trends proposed by Stock and Watson (1988), King et al. (1991), Mellander et al.
(1992), and Warne (1993). The methodology consists of estimating a VAR model with common
stochastic trends. To this end, cointegration restrictions are identi…ed and imposed with the aim
of identifying the relevant parameters and shocks, to then perform a variance decomposition based
on permanent and transitory, as well as foreign and domestic, shocks.
Consider yt a vector that contains n variables, expressed by way of a V AR(k) as follows:
yt =
k
X
i yt i
+ Dt + t ;
(1)
i=1
where Dt refers to the deterministic part and t to the Gaussian residuals, which follow a Normal
distribution with mean 0 and variance . The equation (1) may also be expressed as a vector error
correction model (VECM), as follows:
yt =
yt
1
k 1
X
i
yt
i
+ Dt + t ;
(2)
i=1
P
P
where = I + ki=1 i , i = kj=i+1 j . Cointegration exists when the matrix has reduced
0
range; that is, r < n and in this case, the matrix can be factorized and expressed as =
;
where and are full rank matrices n r. On the one hand, the matrix contains the adjustment coe¢ cients and, on the other hand, the matrix contains the cointegration vectors. To …nd
the cointegration vectors, …rst the vectors of dimension n 1 of residuals r0t and r1t of auxiliary
residuals are obtained; that is, regressions of yt and yt 1 taking into account Dt and the lagged
variables, represented
by f yt 1 ; yt 2 ; :::; yt k+1 g. Thus, the product of residuals is obtained:
PT
0
Sij = (1=T ) i=1 rit rjt , going on to resolve the problem of eigenvalues j S11 S10 S001 S01 j = 0.
From there, the eigenvalues b1 b2 ::: bn and the cointegration values corresponding to the
b b
b
corresponding cointegration
Pnvectors 1 ; 2b; :::; n are obtained. The proposed statistics are denominated: (i) T raza = T i=r+1 log(1
i ), which veri…es the null hypothesis that cointegration
does not exist, r = 0, versus the hypothesis that at least one vector exists, r > 0, and so on until the
3
null hypothesis is not rejected; and (ii) the max = T log(1 bi ) that contrasts the null hypothesis
that r = r0 , in contrast to the alternative hypothesis that r = r0 + 1 where r0 = 0; 1; :::; n 1, until
the null hypothesis is not rejected.
Given the existence of n variables in the model, n r = g stochastic trends exist. The basic
idea behind the common trends representation is that if a system of non-stationary variables exists,
where there are one or more cointegration relationships, it can be decomposed into two components:
one of a permanent nature, and another of a transitory nature. In this case it is said that the system
admits a common trends representation. Mellander et al. (1992) and Warne (1993) represent the
model as follows; see also Stock and Watson (1988):
yt = y0 + =
t
+ (L) t ;
(3)
where L is the lag operator and the vector t satis…es E[ t ] = 0, E[ t 0t ] = In , E[ t 0j ] = 0 for
P1
i
s 6= j and (L) t =
(L) t jointly
i=0 i L . Moreover, y0 is assumed to be stationary and
stationary. On the other hand, = t is the trend component, with = being of dimension n g. Let
t be a random walk vector of dimension g, represented by: t = + t 1 + t where innovation t
is white noise with E[ t ] = 0 and E[ t 0t ] = Ig . Given that a common stochastic trends model is
made up of a vector of trends and a vector of stationary variables, both unobservable as individual
factors, the vector yt can be represented by way of the sum of one permanent component
P and one
transitory component: yt = ytp + yts , where ytp = =[ + t 1 + t ] = =[ 0 + t + ti=1 i ] and,
on the other hand, yts = y0 + (L) t . Given that the number of common trends is g < n, there
are r = n g linearly independent cointegrated vectors, which are orthogonal to the columns of
the matrix =; that is 0 = = 0, zt = 0 yt is stationary and conforms with 0 ytp = 0. The trends
include stochastic elements that are consistent with the idea that some shocks in the economy are
permanent, represented by ytp , while yts contains those shocks that are transitory3 .
The estimation of the common trends model starts o¤ with the assumption that fyt g follows a
VAR generating process of order k, represented by: A(L)yt = +Pt with f t g being white noise,
p
i
while the polynomial matrix of an order of n n A(L) = In
i=1 Ai L satis…es det[A(L)] =
0, if jLj > 1 or L = 1. If the series that comprises fyt g are …rst-order cointegrated; that is,
fyt g CI(1; 1), with r cointegration vectors, Granger’s Representation Theorem establishes that
0
rank[A(1)] = r, A(1) =
. In this sense, the previous VAR can be rewritten as a vector of
error correction model (VECM) represented by A(L) yt P
=
zt 1 + t , where zt = 0 yt is
k 1
i
jointly stationary and the polynomial matrix A(L) = In
i=1 Ai L is related to A(L) through
Pk
Ai =
j=i+1 Aj . On the other hand, according to Wold’s Representation Theorem, there is a
moving average representation
for the previous equation: yt = + C(L) t , where f yt g is jointly
P
j
stationary, C(L) = In + 1
C
j=0 j L and, given that C(1) 6= 0, fyt g is not stationary. Moreover, it is
found that fyt g CI(1; 1), C(1) has a rank n r and 0 C(1) = 0. In this sense, the representation
of moving averages together with the cointegration process imply the existence of an unrestricted
VAR representation and of the VECM representation4 . For the estimation of the common trend
model parameters, an analysis is conducted of whether fyt g exhibits g common stochastic trends,
3
The restrictions of cointegration determine the number of common trends and the way that the yt vector is
related to these.
4
0
As an equivalent, P
Johansen (1991) holds that if fyt g is generated by (2), A(1) =
, and the matrix of dimension
p
0
(n r) (n r): ? [ j=1 jAj ] ? is not singular, so fyt g CI(1; 1) with r vectors of cointegration, leading him to
conclude that there is a representation of moving averages. The necessity for nonsingularity of the matrix mentioned
is important so as to eliminate the possibility that fyt g is integrated of an order greater than one.
4
which allows the matrix = to be rewritten as the product of the matrices = = =0 z where =0 is a
n g matrix of known parameters, ensuring that 0 =0 = 0 and that the trend innovations have an
economic interpretation.
In summary, the common trends model given by (3) has g independent stochastic trends ( t ).
This vector t of dimension g has drifts and innovations t . These innovations have a permanent
e¤ect on the stochastic trends, with the result that they are known as permanent shocks. The
matrix = determines the long-term e¤ect of these shocks on the variables yt . Just as occurs with
the structural vector autoregression model, an important aspect in this context is the identi…cation
of the structural shocks; that is, restrictions need to be imposed on the matrix =. In practice,
as in this paper, the approach taken is to estimate the cointegration vectors , identify them
in accordance with economic theory, and then identify and estimate the matrix =. An evident
restriction is the fact that the permanent shocks are orthogonal to the transitory shocks. In our
case we need a further restriction to identify =. In the following section we detail the restrictions
imposed on the cointegration space and the common trends.
3
Results
3.1
The Data
Following on from Mellander et al. (1992), we de…ne the vector of variables as yt = [ytdt ; ygdpt ; yct ;
yprit ; ypubt ]0 , where ytdt denotes the terms of trade (export price index/index price index), ygdpt is
the real gross domestic product, yct is real consumption, yprit is real private investment, and ypubt
is real public investment. All variables are in logarithms. This is a small open economy model
that is an extension of the closed economy model of King et al. (1991), and is very similar to that
proposed and analyzed by Castillo and Salas (2010). The di¤erence is that we include the total
investment divided into its two components (private and public), with which our model factors in
…ve variables. We would judge the decomposition of total investment to be important not only
due to the di¤erent behaviors of these variables, but also because the role of private investment
has been instrumental to the growth of the Peruvian economy through the favorable development
of the terms of trade. To the best of our knowledge, there is no empirical analysis that analyses
the two investment types separately. This analysis has been performed on the quarterly database
published by the Central Reserve Bank of Peru for the period 1992:01-2007:045;6 .
Figure 1 shows the evolutions of the variables and Figure 2 shows the respective growth rates.
The GDP variable shows a clear positive trend from the start to the end of the period analyzed. It
should be mentioned that the start of the analysis coincides with the beginning of the term of o¢ ce
of the then President Fujimori. The beginning of the economic growth of the Peruvian economy
was favored by two important events: the end of hyperin‡ation period and the capture of the
leader of the terrorist movement called Shining Path. These facts were crucial to form a context of
trust and momentum of a series of structural reforms characterized by a drastic shift to neoliberal
policies, …scal correction, an aggressive privatization of …rms and trade liberalization. This growth
bene…tted even more from the rise in international metal prices, which meant that in 2008, a
period of international crisis, the GDP grew by 9%. Mining is the main activity that underpins the
5
The series were seasonal adjusted using the tramo/seats method developed by Gómez and Maravall (1996). This
procedure was carried out for all variables with the exception of the terms of trade.
6
Castillo and Salas (2010) perform their analysis from 1996 up to 2007.
5
Peruvian primary export model, as an increase in the price of metals raises an economy’s income
levels.
Private investment also shows a positive trend throughout the period of analysis. The price
growth of raw materials implies an increase in aggregate demand component being private investment one of the most favored. This can be termed as the channel of the private investment.
Furthermore, the privatization process that took place in Peru from 1991 constituted a key incentive for the development of this variable. Moreover, the economy’s indicators showed a clear
improvement, as did the economy itself, which encouraged private investment. Gonzáles de Olarte
(1996) holds that the adjustment policy rolled out from the start of the 1990s, as well as the reforms
of the institutional and legal framework, created a new platform for private investment. Equal opportunities for investors, the free movement of factors, goods and services within and outside the
country, legal stability, tax stability agreements for foreign investment, limitations on the economic
role of the state, and free competition constituted a new framework to promote private investment.
It should be noted that throughout the period of study, private investment was largely drawn from
foreign capital. From 1990, the minimization of uncertainty regarding the country’s course in response to economic, political and social instability allowed investors’trust to be regained, and had
positive repercussions on their interest in investing in Peru.
As with the product and private investment, public investment follows a positive trend during
the period under analysis. In fact, the other channel of aggregate demand is on the side of public
investment. This component increases when commodity prices increases because tax revenues
increase. This is evident since tax revenues are a function of the earnings of the companies which
are subject to the income tax and it is higher when the price of raw materials is higher. On the
other hand, public investment has the peculiarity of being closely linked to the decision-making
of the government of the day relating to the country’s course, its priorities, budget level, and the
…scal policy it undertakes to pursue during its term. Insofar as terms of trade shocks are favorable,
in the long term, both private and public investment will increase7 . This result is consistent with
the responsible policy-making decision to increase government spending in the face of a permanent
rise in the terms of trade, whether due to a rise in the price of export goods or a fall in the price
of import goods, or to maintaining a policy of saving if the shock is perceived to be transitory.
Private investment grew by 21.0% during the …nal quarter of 2007, after maintaining a path of
continuous growth from the start of 2002. Moreover, …gures higher than 40.0% in the growth of this
variable in the mid-1990s are observed. In this study’s total period of analysis, private investment
went from 6.2% growth over the period 1992-2000 to 10.5% over the period 2001-2007.
On the other hand, public investment achieved 26.8% growth in the …nal quarter of 2007, as
a result of the set of executions carried out by local governments and the central government.
Nonetheless, in comparison with private investment, its development posted isolated periods of
high growth, represented by …gures in excess of 20%, followed by periods of sudden slowdown.
Moreover, this latter variable posted a decrease in its average growth from the period 1992-2000 to
the period 2001-2007, dropping from 6.1% to 2.9%.
Private investment retained an important role in the dynamics of Peru’s economy over the
7
It is important to mention that there may be other channels through which the terms of trade can a¤ect the
economy. One is through the balance of payments, i.e. through availability of international reserves. The other
mechanism is through capital ‡ows although the literature is not clear whether there is independence between these
two variables. However, it is important to note that these mechanisms are not modeled in this paper but can be
signi…cant.
6
period 1992-2007. In comparison with public investment, private sector investment has accounted
for, on average, more than 17% of GDP, while public investment has only amounted to 4%. In
addition, over the same period of analysis private investment has constituted more than 70% of the
total investment in the country. Moreover, the analysis of private investment from 2000 to 2007
records a …gure in excess of 80% with respect to the total value of investment in the country, as
well as a growing trend of sustained growth, increasing from an average quarterly growth of 1.5%
in the years 2000-2004 to 18% over the period 2005-2007.
Finally, the terms of trade series presents a positive trend from 2000, from a previous stage of
slowdown and with the exception of a sharp fall in this series in 2007. It is worth mentioning that
in 2007 the symptoms of the …nancial crisis occasioned in the United States began to be felt, which
caused an imbalance in the country’s import and export prices.
3.2
Cointegration
As a preliminary stage to the application of the methodology formulated in the previous section,
the stationarity of the variables is examined using the ADF statistic proposed by Said and Dickey
(1984), the statistics ADF GLS , PTGLS and M GLS put forward by Elliott, Rothenberg and Stock
(1996), and Ng and Perron (2001), respectively. For the selection of lags, the Modi…ed Akaike
Information Criteria (MAIC), proposed by Ng and Perron (2001), is employed. All results support
non-rejection of the null hypothesis of the unit root in all of the model’s variables8 .
After an examination of the lag structure, a VAR(6) is estimated, with good results related to the
behavior of the residuals9 . Table 1 shows the results obtained from Johansen’s cointegration tests
(1988, 1995). According to the Trace statistic and the max , the results show three cointegration
relationships; restrictions were imposed to identify these three relationships. The …rst equation
normalized with respect to ygdpt , while the second and third equations normalized with respect to
yprit . The statistic calculated with the restrictions is equal to 8.369, which is less than the critical
value at a level of signi…cance of 5% (p-value=0.592). In consequence, the long-term relationships
between the variables proposed are not rejected. In this regard, the three equations have the
following representation:
0
2
0.0
6
6
yt = 6 -1.5
4
0.0
1.0
-1.16
0.0
-0.4
0.0
1.0
-0.80
0.0
1.0
2
ytdt
36
6
6 y
0.0
7 6 gdpt
76
0.0 7 6 yct
56
6
-1.0 6 yprit
4
ypubt
3
7
7
7
7
7
7:
7
7
7
5
(4)
The …rst cointegration relationship displays a positive relationship of ygdpt with respect to yct .
On the other hand, the second cointegration relationship assumes a positive relationship of yprit
together with ygdpt and ytdt . Finally, the third equation again shows a positive relationship, but
of a greater magnitude, between yprit and ygdpt . In this …nal relationship, the terms of trade are
not present but ypubt is. The investment made by the public sector shows a positive relationship
8
Results available on request.
A variable dummy is used for the fall in consumption (1998:4). All results of the evaluation of residuals are
available.
9
7
with respect to private investment; that is, a complementary relationship between both investments
is observed. This complementary relationship may be justi…ed for di¤erent reasons. On the one
hand, the government’s decision to execute an economic policy in which an investment plan for
infrastructure is put forward, means that the interested parties regard it attractive to place their
investments at the same time as those of the government. Moreover, the state’s economic impulse
is perceived as a guarantee over the country’s political stability. On the other hand, the investment
made by the government turns out to be an incentive to private investment, as new projects are
launched that had not previously been considered due to high risks of economic loss in the event of
failure. Moreover, the increase in the promotion of joint projects -known as public-private projectsin the last three years, in which the state and the private sector participate, supports this result.
Figure 3 shows the three cointegration relationships speci…ed in (4). Some further restrictions
on the adjustment coe¢ cients in the cointegration model are imposed. The …rst set of restrictions
are associated with the fact that the terms of trade do not respond to domestic imbalances. These
restrictions are commonly used. The other restriction is that the yct does not respond to the
imbalance of the second cointegration relationship. Given that this relationship is made up of
ydgpt , yprit and ytdt , this assumption seems reasonable.
3.3
Common Trends
In addition to the restrictions imposed thus far, we need a further restriction in order to identify
the matrix = given that the model presents two common trends (n r = g = 5 3 = 2). It is
assumed that the shocks to one of the trends (called shocks to the domestic trend) does not have
long-term e¤ects on the terms of trade. This is equivalent to saying =12=0 = 0, which allows us to
identify the stochastic trends as one foreign ( f;t ) and the other domestic ( d;t ).
The results of the matrix for the estimation of the common trend parameters is represented by:
2
3
3
2
0.051
0.000
6
7
y
td
6
7
t
7
6
6 (0.024) (0.000) 7
7
6
6
7
7
6
6
7
7
6
6
0.022
0.004 7
7
6
6
7
6 ygdp 7
6
7
t 7
6
6
(0.011) (0.001) 7
7
6
6
72
7
6
3
6
7
7
6
6
7 b
7
6
0.019
0.004
6
7 4 f;t 5 b
7
6 y
+ (L) t :
(5)
7
6 ct 7 = b + 6
6
7 b
7
6
(0.009)
0.001)
d;t
6
7
7
6
6
7
7
6
6
7
7
6
0.002 7
6 0.084
7
6 y
6
7
6 prit 7
6
7
7
6
(0.041)
(0.000)
6
7
7
6
6
7
7
6
6
7
5
4
-0.002 7
6 0.067
4
5
ypubt
(0.032) (0.000)
The estimation of the parameters shows that both trends, foreign ( f ) and domestic ( d ), possess
long-term positive e¤ects on all of the variables analyzed, except for the e¤ect of domestic shock on
ypubt and the null e¤ect of the domestic trend on ytdt (by assumption). All coe¢ cients estimated
have a level of signi…cance of 5%.
8
In the long-term, the response of ytdt to a deviation of the permanent external shock is positive
and equal to 5.1%. On the other hand, because of the assumption of the model as regards the small
size of the economy, the common trend has no in‡uence on this variable, so the e¤ect of the domestic
trend on ytdt is null. The e¤ect of a terms of trade shock on ygdpt is 2.2%. Likewise, yct is positively
altered by almost the same extent (1.9%) in response to a deviation of foreign shock. This result
is also found by Castillo and Salas (2010) and supports economic theory, which suggests the small
change to the savings rate. For the case of the e¤ect of the foreign trend on yprit , the result shows
a positive change of 8.5%. The e¤ect is higher than that reported by the domestic trend, which
is very close to zero (0.2%). The price growth of raw materials implies an increase in aggregate
demand component being private investment one of the most favored. This can be termed as the
channel private investment. Finally, the matrix also indicates a positive e¤ect of 6.7% of the foreign
trend for the case of ypubt . This is the other channel of aggregate demand (on the side of public
investment). This component increases when commodity prices move up because increase in tax
revenues. This is because tax revenues are a function of the earnings of the companies which are
subject to the income tax and it is higher when the price of raw materials is higher. Nonetheless,
a disturbance in the domestic trend on ypubt possesses a negative e¤ect equal to 0.2%. The returns
of yprit are usually quantitatively greater and the incentives of public policy-makers are not usually
well-de…ned, which supports the result that public investment is displaced by private investment
in those years in which the total productivity of factors has increased, all other things being equal.
The matrix of the parameters of the common trends indicates that the parameter found in this
estimation for the case of the e¤ect of the domestic trend on both investments is exactly the same,
but with opposite signs. That is, in the long term, the productivity shock explains, to the same
degree, the change in private and public investment, but with opposite signs10 . In summary, what is
found is a strong long-term impact of foreign factors on private and public investment, respectively.
Figure 4 shows the transitory components obtained as the di¤erence between the variable observed and its permanent component. Meanwhile, Figure 5 shows the permanent components,
together with the development of each of the variables in the system. The variables ygdpt and yct
have very closely followed the behavior of their permanent components. Nonetheless, a deviation
of both series during the …ve-year period of study is observed, leaving both below their permanent
component, which indicates the presence of transitory factors. Likewise, the private investment
series displays behavior similar to the path of its permanent component during the period of analysis, and a deviation over the last …ve years. Moreover, another similar …ve-year period is observed,
1996-2001, in which the behavior of private investment is above its permanent component, which
again indicates the presence of transitory factors during both …ve-year periods in mention. For the
case of the terms of trade, it can be seen that the accelerated growth of the series, from 2004 to
2006, is clearly explained by the permanent component of the series. Finally, public investment
posts the largest deviations, whether above or below, with respect to its permanent component,
with which a greater presence of transitory factors can be inferred.
10
Another plausible explanation for the opposite sign between the two types of investment is the fact that since
1992 and particularly since 1994 strong structural reforms among which was a major privatization of state enterprises
were established. This caused a shift or reduction in public investment.
9
3.4
Variance Decomposition
Table 2 displays the variance decomposition of di¤erent time horizons, ranging from the short term
to a medium-term horizon (horizons h = 1; 4; 8; 12; 20; 40). It can be seen that the set of transitory
(and domestic) shocks under any time horizon minimally explains the behavior of the terms of
trade, which is assumed and consistent with the behavior of a small open economy. With respect
to the other variables, the transitory components, even if they possess signi…cant weight, continue
to display decreasing behavior as the medium-term is reached.
In the very short-term (less than one year), a signi…cant proportion of private consumption is
explained by the set of transitory shocks, determining 68% of its total variability. Nonetheless, the
proportion decreases signi…cantly from the year of analysis. In the same way, the e¤ect of the set
of transitory shocks on the product shows a value greater than 20% in the short-term, dropping to
a value close to zero in the medium-term. Likewise, the e¤ects of the set of transitory shocks on
both investments are signi…cant at the start of the period analyzed, and are later reduced in the
medium-term. It is worth noting that the reduction of the proportion explained by the transitory
public investment shocks displays gradual behavior, remaining at 10% after 20 quarters, while the
same e¤ect on private investment loses signi…cance after eight quarters, which represents less than
half of the time elapsed in the case of public investment, dropping to less than 10% after having
accounted for almost 40% of this investment over the …rst four quarters.
With respect to the impact of permanent shocks on the variables studied, Table 2 shows that
permanent foreign shocks account for almost all terms of trade variability. The percentage value
of the external shock on the terms of trade ‡uctuates between 97.1% and 99.5% in the short- and
medium-term. This shows that a ‡uctuation in the terms of trade in the short- and medium-term
is primarily explained by a foreign shock, which makes sense in the case of a small open economy
such as Peru’s.
As regards private consumption, it is seen that the permanent foreign shock went from being
insigni…cant (below 5%) in the beginning, to taking a value in excess of 50% after four quarters
and …nally reaching 96% in the medium-term.
For the case of the product, the permanent foreign shock displays almost null e¤ect in the …rst
quarter of analysis. Nonetheless, a signi…cant increase is observed in the importance of the e¤ect in
the determination of the product during subsequent periods, accounting for 51% of its variability
during the short term, and going on to 95% in the medium term.
Finally, the e¤ect of permanent shocks on the behavior of disaggregated investment in public
and private investment can be seen. Foreign shock has a leading role in determining both levels
of investment. Speci…cally, the proportion of this e¤ect is relatively high in the case of private
investment. The value observed for this latter investment is greater than 90% from the eighth
quarter, while the e¤ect of permanent foreign shock on public investment only takes a similar value
several quarters later. Domestic shock, for its part, has a no role in either level of investment in
the short- and medium-term. In particular, this is signi…cant, with 13% for private investment in
the very short–term, but later loses signi…cance in the medium-term.
Table 3 shows the long-term variance decomposition. The components that comprise the Table
correspond to the response to the variables analyzed given some of the permanent shocks encountered. In this case, only permanent shocks are taken, as in the long term the ‡uctuation of the
variables are no longer explained by transitory shocks. The results found are similar to those presented by Castillo and Salas (2010). The estimation shows that in the long term, the ‡uctuation
10
of the product, private consumption, and public and private investment is explained, almost entirely, by foreign shocks. In the case of all variables, the foreign shock on the ‡uctuation of these
corresponds to more than 95%. Public and private investments post a bigger role for foreign shock,
followed by the product and consumption. This had already been observed in the coe¢ cients of
the matrix =.
On the other hand, Table 3 points out that, in general, the long-term impact of domestic shock
on the variables is insigni…cant, attaining a participation of only 3% in the case of the product and
private consumption. The results found are not surprising for an emerging economy such as Peru’s.
The weak development of internal industry and, thus, the dependence on primary activity render
the country more vulnerable to external shocks. The economy’s revenues have the production of
primary goods as their main source; as a result, the variations of the terms of trade maintain
a direct e¤ect on the country’s economic vulnerability. Even more so, the Peruvian economy is
characterized by a high concentration of raw material exports, which means that a variation in the
prices of its goods causes instability on the revenues received.
3.5
Historical Decomposition
Finally, we perform a decomposition of the historical development of the Peruvian economy’s main
macroeconomic variables. Unlike the analysis of the variance decomposition, which is based on
the estimation of the shocks, the historical decomposition is based on the current performances
of the shocks. In this regard, the historical decomposition enables an alternative measurement
of the relative importance of foreign and domestic shocks in the development of the estimated
permanent components of each variable (ygdpt ; yct ; yprit ; ypubt )11 . The analysis was carried out for
the entire sample, and the period of study was also disaggregated into two sub-periods: 1994-2000
and 2001-2007, in which the second period was characterized by favorable terms of trade for Peru.
The results of this analysis, presented in Table 4, show an increase in the trend product between
the 1990s and 2000s, from 2.7% to 4.9%, respectively. Its development is explained primarily by
the development in the terms of trade, which went from contributing -0.4% of the trend product in
the 1990s to 1.9% in the 2000s. The domestic trend factors, which re‡ect the development of total
domestic factor productivity, remained stable in the two sub-periods, at around 3%.
The analysis for the case of potential consumption shows results similar to those obtained for
the product. Thus, it is estimated that the trend growth of consumption rose from 2.3% in the
1990s to 4.3% in the 2000s. As in the case of the product, the change in the terms of trade, from
-0.3% in the 1990s to 1.6% in the 2000s, was a determinant for the trend increase in consumption.
Moreover, the domestic trend factors stayed at around 2.7%.
In the case of the trend growth in private investment, this grew from 0.3% in the 1990s to 8.5%
in the 2000s. The increase of 8.2 percentage points from one sub-period to another is primarily
explained by a trend growth associated with the development of the terms of trade (7.3%), while
the participation of the domestic factors remained at around 1.3%. It is worth mentioning that
the participation of external factors in the private consumption trend development amounts to
approximately one-…fth of such participation in the private investment trend development, while in
the case of the participation of domestic factors, the relationship between both is almost double.
Unlike the other variables, the public investment trend development is the only series in which
the domestic factors have a negative e¤ect, increasing from -1.0% in the 1990s to -1.2% in the
11
This will be known as the potential components of the variables.
11
2000s. That is, in those years in which the total productivity of factors increased (all other things
being equal), public investment fell. Nonetheless, the trend growth of public investment, as in the
case of the other variables, is primarily explained by external factors, rising from a participation of
-0.9% in the 1990s to 5.8% in the 2000s.
On the other hand, it can be seen that the contribution of domestic factors remained stable
during the second sub-period, while the contribution of foreign factors to the development of all
variables increased signi…cantly for the second period. It is worth mentioning that the contribution
of foreign factors was more pronounced for the case of public and private investment, with increases
from -0.9% to 5.8% and from -1.1% to 7.3%, respectively.
The analysis of historical decomposition also allows the observation that the average participation of foreign factors in the development of all variables went from being negative over the period
1994-2000 to positive over the period 2001-2007. That is, during the …rst sub-period, foreign factors -rather than contributing positively to the growth of the variables- contributed negatively on
average, stalling the economy’s growth to a certain degree. Nonetheless, during the second subperiod, on average the external factors contributed positively to the growth of economic variables,
accentuating their contribution to the development of both investments. In this sense, the results
highlight that the volatility of the variables studied is highly related to external factors.
The greatest positive contribution of the terms of trade shocks to the permanent component of
the product, consumption, private investment and public investment was posted in 2006, a period
in which Peru constituted a favorable stage for foreign trade. On the other hand, the greatest
negative contribution of foreign shocks to all variables mentioned was recorded one year later, in
2007, the period that marked the start of the international subprime crisis.
The observations of historical decomposition obtained may seem contradictory in comparison
to those obtained of the variance decomposition. In e¤ect, Table 4 suggests that domestic factors
have been the most important in explaining the average growth rate in the potential of each of the
variables under analysis. Nonetheless, Table 4 also suggests that the volatility of the potential of
the four variables is dependent on foreign factors. In other words, the potential level of the four
domestic variables is highly dependent on the external factors in the sense that their volatilities
depend much on the movements in the terms of trade. The domestic component, we can a¢ rm,
has been more stable.
4
Conclusions
This document has four objectives. Firstly, to verify the existence of long term relationships
between the group of variables analyzed. Secondly, to analyze the role of private investment and
public investment, as well as the role of terms of trade in the economic ‡uctuations of the Peruvian
economy. Thirdly, to identify the domestic and foreign shocks, as well as the degree of importance
of both in the economy’s ‡uctuations. Finally, to identify the role and impact of permanent and
transitory shocks in the academic ‡uctuations of an emerging economy such as Peru’s.
The results indicate the existence of three long term relationships between the set of variables
and two stochastic trends, which are comprised of domestic and foreign shocks. With respect
to the relationship existing between private and public investment, the analysis of cointegration
proposes a complimentary e¤ect between both, which is corroborated with the results obtained
from the analysis of common stochastic trends in response to the presence of permanent foreign
shocks (terms of trade). That is, insofar as the shocks in the terms of trade are favorable, in the
12
long term, both private and public investment will increase, which is consistent with responsible
policy-making behavior on increasing …scal spending in the face of a permanent increase in the
terms of trade and, on the other hand, maintaining a savings policy if the shock is perceived to be
transitory. Likewise, in the face of a permanent fall in the terms of trade, both private and public
investment will also fall to the same extent. Nonetheless, a domestic shock (productivity) suggests
a substitution relationship (or crowding out e¤ect) between both investments, which is consistent
with the observation in the period of structural reforms (1990-1996) in the country, in which private
investment played an important role in the growth of the economy, while the role of the state was
historically reduced.
Moreover, the analysis of stochastic trends indicates the role of permanent shocks in the mediumand long-term on all variables analyzed, while the transitory shocks explain, to a very limited
degree, the behavior of the variables in the very short-term. In this sense, it is concluded that
the high growth in the main macroeconomic principles of an emerging country such as Peru, in
the scenario of a small open economy, is signi…cantly linked with external factors, such as the rise
in the prices of raw materials. Complimentarily, the decomposition of the historical development
of the main variables is analyzed, for the periods 1994-2000 and 2001-2007, as another way of
measuring the relative importance of external factors versus domestic factors in the development
of each variable. The results also indicate the high sensitivity of the Peruvian economy’s main
macroeconomic variables in the face of external factors.
The study shows that the behavior of all variables analyzed is primarily explained by the
permanent foreign shocks in the long-term. This result took on great importance in the behavior
of the two investment series, led by private investment. In this sense, the e¤ect of the permanent
domestic shocks in the behavior of the variables studied is insigni…cant. This indicates the Peruvian
economy’s high vulnerability to external factors due to greater international economic exposure,
which is exacerbated by the relatively small size of the domestic economy compared to the worldwide
economy.
The conclusions set out here show the e¤ects experienced by a small and open economy characterized by the production of primary goods as the main economic activity, which are directly related
to the international market and where a price variation has direct repercussions on the economy’s
terms of trade. Moreover, the concentration of exports in the Peruvian economy is another key
factor in understanding the depth of economic vulnerability with respect to external factors. A
small concentration would enable greater stability in the …nal revenues received by the economy
and, in turn, the country’s economic activity would be less vulnerable.
Policy-makers face a considerable task based on the desire to cushion the e¤ects of foreign shocks
on the economy. The monetary policy, exercised by the each country’s Central Bank, is an important
tool in tackling the adverse consequences of external shocks. The role of each Central Bank in the
absorption or ampli…cation of shocks arising from outside the country, directly intervening through
its policy instruments and indirectly on exchange rate volatility, will be decisive in the rise of fall
of internal economic vulnerability.
The results are consistent with the favorable scenario of the terms of trade and its e¤ects on a
small open economy such as Peru’s. The best external conditions have been conducive to signi…cant
growth in private investment in mining and the exportation of raw materials associated with that
sector. At present, the terms of trade are unfavorable to the Peruvian economy and the e¤ects on
the main macroeconomic variables has quickly gained in prominence through a reduction in various
private investment projects related to mining, as well as problems in executing private investment.
13
All of this has contributed to the Peruvian economy leaving behind high growth levels (6% to 8%
annual) to return to growth rates of 2% or less.
References
[1] Aguiar, M. and G. Gopinath (2007), “Emerging Market Business Cycles: The Cycle is the
Trend”, Journal of Political Economy 115, 69-102.
[2] Ahmed, S. and R. Murphy (1994), “Money, Output and Real Business Cycles in a Small Open
Economy”, Canadian Journal of Economics 27, 982-993.
[3] Becker, T. and P. Mauro (2006), “Output Drops and the Shocks that Matter,” IMF Working
Papers 06/172.
[4] Blattman, C., J. Hwang and J. Williamson (2004), “The Impact of the Terms of Trade on
Economic Development in the Periphery, 1870-1939: Volatility and Secular Change,” NBER
Working Papers 10600.
[5] Canova, F. (2005), The Transmission of US Shocks to Latin America, Journal of Applied
Econometrics 20(2), 229 - 251.
[6] Castillo, P. and J. Salas (2010), “The terms of trade as drivers of economic ‡uctuations in developing economies: an empirical study”, Unpublished manuscript, Banco Central de Reserva
del Perú.
[7] Dancourt, O., W. Mendoza, and L. Vilcapoma (1997), “Fluctuaciones económicas shocks externos, Perú 1950- 1996”, Departamento de Economía, Ponti…cia Universidad Católica del
Perú, Documento de Trabajo 135.
[8] Easterly, W., R. Islam and J. Stiglitz (2001), “Shaken and Stirred: Explaining Growth Volatility,” in: Pleskovic, B. and Stern, N. (eds.), Annual World Bank Conference on Development
Economics 2000, The World Bank, 191-211.
[9] Elliott, G., T. J. Rothenberg and J. H. Stock (1996), “E¢ cient tests for an autoregressive unit
root”, Econometrica 55, 143-59.
[10] Gómez, V. and A. Maravall (1996), “Programs TRAMO (Time series Regression with Arima
noise, Missing observations and Outliers) and SEATS (Signal Extraction in Arima Time Series)”. Servicios de Estudios del Banco de España, Documento de trabajo 9628.
[11] Gonzáles de Olarte, E. (1996), “Inversión privada, crecimiento y ajuste estructural en el Perú,
1950- 1995”, IEP, Documento de Trabajo 81.
[12] Ho¤maister, A. and J. Roldós (1997), “Are business cycles di¤erent in Asia and Latin America?”, IMF Working Paper 9.
[13] Ho¤maister, A., J. Roldós and P. Wickman (1998), “Macroeconomic Fluctuations in SubSaharan Africa”, IMF Sta¤ Papers 45(1).
14
[14] International Monetary Fund (2008), World Economic Outlook, April.
[15] International Monetary Fund (2014a), “Peru. Selected Issues Paper,” IMF Country Report
14/22.
[16] International Monetary Fund (2014b), “World Economic Outlook. Recovery Strengthens, Remains Uneven,” April.
[17] Izquierdo, A., R. Romero, and E. Talvi, E. (2008), Booms and Busts in Latin America: The
Role of External Factors, IADB Research Department, Working Paper 631.
[18] Johansen, S. (1988), “Statistical analysis of cointegration vectors”, Journal of Economics Dynamics and Control 12, 231-254.
[19] Johansen, S. (1991), “Estimation and hypothesis testing of cointegration vectors in Gaussian
vector autoregressive models,” Econometrica 59, 1551-1580.
[20] Johansen, S. (1995), Likelihood-based inference in cointegrated vector autoregressive models,
Oxford Universit Press, New York.
[21] Justiniano, A. and B. Preston (2008), “Can Structural Small Open Economy Models Account
for the In‡uence of Foreign Disturbances?,” NBER Working Paper 14547.
[22] King, R. ,C. Plosser, J. H. Stock and M. Watson (1991), “Stochastic trends and economic
‡uctuations”, American Economic Review 81, 819-840.
[23] Kose, M. A. (2002), “Explaining Business Cycles in Small Open Economies: How much do
World Prices Matter?”, Journal of International Economics 56, 299-327.
[24] Kose M. A. and R. Riezman (1999), “Trade shocks and macroeconomic ‡uctuations in Africa”,
CESifo, Working paper N 203.
[25] Lubik, T. and Teo, W. (2005), “Do World Shocks Drive Domestic Business Cycles? Some
Evidence from Structural Estimation, Economics,” Working Paper Archive 522, The Johns
Hopkins University, Department of Economics.
[26] Mellander, E., A. Vredin and A. Warne (1992), “Stochastic trends and economic ‡uctuations
in a small open economy”, Journal of Applied Econometrics 7(4), 369-394.
[27] Mendoza, E. (1995), “The terms of trade, the real exchange rate, and economic ‡uctuations”,
International Economic Review 36, 101-137.
[28] Misas, M., E. López and D. Vásquez (2003), “Tendencias estocásticas comunes y ‡uctuaciones
en la economía colombiana 1950-2002”, Borradores de Economía 275, Banco de la República.
[29] Ng, S. and P. Perron (2001), “Lag length selection and the construction of unit root tests with
good size and power”, Econometrica 69, 1519-554.
[30] Obstfeld, M. (1982), Aggregate Spending and the Terms of Trade: Is There a Laursen-Metzler
E¤ect?, The Quarterly Journal of Economics 97(2), 251-270.
15
[31] Said, E. and D. A. Dickey (1984), “Testing for unit roots in autoregressive moving average
models of unknown order”, Biometrika 71, 599-607.
[32] Stock, J. H. and M. Watson (1988), “Testing for common trends”, Journal of the American
Statistical Association 83, 1097-1107.
[33] Svensson, L. and A. Razin (1983), The Terms of Trade and the Current Account: The
Harberger-Laursen-Metzler E¤ect, Journal of Political Economy 91(1), 97-125.
[34] Warne, A. (1993), “A commons trends model: identi…cation, estimation and inference”,
Mimeo, Stockholm School of Economics.
16
Table 1. Cointegration Tests
H0
value
critical value at 5%
p-values
Traza (5%)
r=0
146.60
69.82
0.000
r=1
79.64
47.86
0.000
r=2
37.62
29.80
0.005
r=3
13.82
15.50
0.088
r=4
2.10
3.84
0.147
max (5%)
r=0
66.96
33.88
0.000
r=1
42.02
27.58
0.000
r=2
23.81
21.13
0.021
r=3
11.72
14.26
0.122
r=4
2.10
3.84
0.147
T-1
Table 2. Decomposition of Variance
Variable
ytdt
Innovation
f
d
transitory
ygdpt
f
d
transitory
yct
f
d
transitory
yprit
f
d
transitory
ypubt
f
d
transitory
h=1
h=4
h=8
h = 12
h = 20
h = 40
0.971
0.964
0.980
0.985
0.991
0.995
(0.076)
(0.036)
(0.024)
(0.019)
(0.013)
(0.007)
0.004
0.005
0.002
0.002
0.001
0.001
(0.064)
(0.022)
(0.012)
(0.013)
(0.005)
(0.003)
0.026
0.031
0.018
0.013
0.008
0.004
(0.046)
(0.028)
(0.019)
(0.012)
(0.010)
(0.005)
0.001
0.507
0.834
0.914
0.942
0.955
(0.018)
(0.277)
(0.321)
(0.233)
(0.214)
(0.212)
0.646
0.278
0.104
0.058
0.044
0.039
(0.941)
(0.301)
(0.291)
(0.221)
(0.211)
(0.210)
0.353
0.215
0.062
0.028
0.013
0.006
(0.939)
(0.109)
(0.055)
(0.025)
(0.012)
(0.007)
0.021
0.579
0.845
0.916
0.945
0.957
(1.006)
(0.943)
(0.595)
(0.401)
(0.311)
(0.277)
0.301
0.304
0.129
0.072
0.049
0.040
(0.400)
(0.885)
(0.563)
(0.384)
(0.303)
(0.273)
0.677
0.117
0.026
0.011
0.005
0.002
(0.656)
(0.159)
(0.043)
(0.020)
(0.009)
(0.004)
0.032
0.534
0.900
0.964
0.985
0.993
(0.053)
(0.212)
(0.063)
(0.026)
(0.023)
(0.002)
0.134
0.056
0.012
0.004
0.002
0.001
(0.143)
(0.079)
(0.017)
(0.009)
(0.017)
(0.019)
0.834
0.410
0.088
0.031
0.013
0.005
(0.149)
(0.166)
(0.056)
(0.023)
(0.011)
(0.006)
0.122
0.173
0.490
0.719
0.881
0.945
(0.205)
(0.537)
(0.434)
(0.226)
(0.124)
(0.071)
0.016
0.011
0.052
0.046
0.018
0.009
(0.058)
(0.059)
(0.199)
(0.218)
(0.087)
(0.053)
0.863
0.815
0.458
0.235
0.100
0.046
(0.171)
(0.550)
(0.507)
(0.189)
(0.063)
(0.034)
Standard Errors are in parentheses
T-2
Table 3. Long-Run Decomposition of Variance
variable
f
d
ytdt
1.000
0.000
(0.000)
(0.000)
0.966
0.339
(0.036)
(0.036)
0.966
0.034
(0.036)
(0.036)
0.999
0.000
(0.000)
(0.000)
0.999
0.001
(0.001)
(0.001)
ygdpt
yct
yprit
ypubt
Standard Errors are in parentheses
T-3
Table 4. Historical Decomposition (%)
Variable
Total
Foreign
Domestic
1994-2007
Total
Foreign
Domestic
Total
1994-2000
Foreign
Domestic
2001-2007
ygdpt
3.8
0.8
3.1
2.7
-0.4
3.1
4.9
1.9
3.0
yct
3.3
0.6
2.7
2.4
-0.3
2.7
4.3
1.6
2.6
yprit
4.4
3.1
1.3
0.3
-1.1
1.4
8.5
7.3
1.2
ypubt
1.4
2.5
-1.1
-1.9
-0.9
-1.0
4.6
5.8
-1.2
T-4
Output
Consumption
10.8
10.4
10.6
10.2
10.4
10.0
10.2
9.8
10.0
9.6
9.8
9.4
92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07
92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07
Private Investment
Public Investment
9.2
7.6
7.4
8.8
7.2
8.4
7.0
6.8
8.0
6.6
7.6
6.4
92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07
92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07
Terms of Trade
5.0
4.9
4.8
4.7
4.6
4.5
4.4
92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07
Figure 1. Seasonal Adjusted Variables
F-1
60
40
20
0
-20
-40
-60
92
93
94
95
96
97
98
99
00
01
02
03
04
05
Consumption
Private Investment
Public Investment
Output
Figure 2. Growth Rates of Variables
F-2
06
07
.2
.06
.1
.04
.0
.02
-.1
.00
-.2
-.02
-.3
-.4
-.04
93 94 95 96 97 98 99 00 01 02 03 04 05 06 07
93 94 95 96 97 98 99 00 01 02 03 04 05 06 07
Cointegrating Relation 1
Cointegrating Relation 2
.4
.2
.0
-.2
-.4
93 94 95 96 97 98 99 00 01 02 03 04 05 06 07
Cointegrating Relation 3
Figure 3. Estimated Cointegrated Relations
F-3
Consumption
Output
.10
.08
.05
.04
.00
.00
-.05
-.04
-.10
-.08
-.15
-.12
94 95 96 97 98 99 00 01 02 03 04 05 06 07
94 95 96 97 98 99 00 01 02 03 04 05 06 07
Private Investment
Public Investment
.4
.6
.2
.4
.0
.2
-.2
.0
-.4
-.2
-.6
-.4
94 95 96 97 98 99 00 01 02 03 04 05 06 07
94 95 96 97 98 99 00 01 02 03 04 05 06 07
Terms of Trade
.15
.10
.05
.00
-.05
-.10
-.15
94 95 96 97 98 99 00 01 02 03 04 05 06 07
Figure 4. Transitory Components
F-4
Output
Consumption
10.8
10.4
10.7
10.3
10.6
10.2
10.5
10.1
10.4
10.0
10.3
9.9
10.2
9.8
10.1
10.0
9.7
94
95
96
97
98
99
00
01
02
Permanent Component
03
04
05
06
07
94
95
96
97
Output
98
99
00
01
02
Permanent Component
Private Investment
03
04
05
06
07
Consumption
Public Investment
9.4
7.5
9.2
7.4
7.3
9.0
7.2
8.8
7.1
8.6
7.0
8.4
6.9
8.2
6.8
8.0
6.7
94
95
96
97
98
99
00
01
02
03
04
05
06
07
94
95
96
97
98
99
Permanent Component
Private Investment
Terms of Trade
5.0
4.9
4.8
4.7
4.6
4.5
4.4
94
00
01
02
Permanent Component
Public Investment
95
96
97
98
99
00
01
02
03
04
05
06
07
Permanent Component
Terms of Trade
Figure 5. Series and Permanent Components
F-5
03
04
05
06
07
Output
Consumption
.16
.12
.12
.08
.08
.04
.04
.00
.00
-.04
-.04
-.08
-.08
94
95
96
97
98
99
00
01
Domestic Factors
02
03
04
05
06
07
94
95
96
97
Foreign Factors
98
99
00
01
Domestic Factors
Private Investiment
02
03
04
05
06
07
Foreign Factors
Public Investment
.5
.4
.4
.3
.3
.2
.2
.1
.1
.0
.0
-.1
-.1
-.2
-.3
-.2
94
95
96
97
98
99
00
Domestic Factors
01
02
03
04
05
06
07
94
Foreign Factors
95
96
97
98
99
00
01
Domestic Factors
Figure 6. Historical Decomposition of Growth Rates
F-6
02
03
04
Foreign Factors
05
06
07
ÚLTIMAS PUBLICACIONES DE LOS PROFESORES
DEL DEPARTAMENTO DE ECONOMÍA
Libros
José Carlos Orihuela y José Ignacio Távara (Edt.)
2014 Pensamiento económico y cambio social: Homenaje Javier Iguíñiz. Lima, Fondo
Editorial, Pontificia Universidad Católica del Perú.
Jorge Rojas
2014 El sistema privado de pensiones en el Perú. Lima, Fondo Editorial, Pontificia
Universidad Católica del Perú.
Carlos Conteras (Edt.)
2014 El Perú desde las aulas de Ciencias Sociales de la PUCP. Lima, Facultad de Ciencias
Sociales, Pontificia Universidad Católica del Perú.
Waldo Mendoza
2014 Macroeconomía intermedia para América Latina. Lima, Fondo Editorial, Pontificia
Universidad Católica del Perú.
Carlos Conteras (Edt.)
2014 Historia Mínima del Perú. México, El Colegio de México.
Ismael Muñoz
2014 Inclusión social: Enfoques, políticas y gestión pública en el Perú. Lima, Fondo
Editorial, Pontificia Universidad Católica del Perú.
Cecilia Garavito
2014 Microeconomía: Consumidores, productores y estructuras de mercado. Lima, Fondo
Editorial, Pontificia Universidad Católica del Perú.
Alfredo Dammert Lira y Raúl García Carpio
2013 La Economía Mundial ¿Hacia dónde vamos? Lima, Fondo Editorial, Pontificia
Universidad Católica del Perú.
Piero Ghezzi y José Gallardo
2013 Qué se puede hacer con el Perú. Ideas para sostener el crecimiento económico en el
largo plazo. Lima, Fondo Editorial de la Pontificia Universidad Católica del Perú y
Fondo Editorial de la Universidad del Pacífico.
Cecilia Garavito e Ismael Muñoz (Eds.)
2012 Empleo y protección social. Lima, Fondo Editorial, Pontificia Universidad Católica del
Perú.
Serie: Documentos de Trabajo
No. 388
“Aplicación de una metodología para el análisis de las desigualdades
socioeconómicas en acceso a servicios de salud y educación en Perú en 20052012”. Edmundo Beteta Obreros y Juan Manuel del Pozo Segura. Diciembre,
2014.
No. 387
“Sobre la naturaleza multidimensional de la pobreza humana: propuesta
conceptual e implementación empírica para el caso peruano”. Jhonatan A.
Clausen Lizarraga y José Luis Flor Toro. Diciembre, 2014.
No. 386
“Inflation Targeting in Peru: The Reasons for the Success”. Oscar Dancourt.
Diciembre, 2014.
No. 385
“An Application of a Short Memory Model With Random Level Shifts to the
Volatility of Latin American Stock Market Returns”. Gabriel Rodríguez y Roxana
Tramontana. Diciembre, 2014.
No. 384
“The New Keynesian Framework for a Small Open Economy with Structural
Breaks: Empirical Evidence from Perú”. Walter Bazán-Palomino y Gabriel
Rodríguez. Noviembre, 2014.
No. 383
“An Application of a Random Level Shifts Model to the Volatility of Peruvian
Stock and Exchange Rates Returns”. Junior Ojeda y Gabriel Rodríguez.
Noviembre, 2014.
No. 382
“Firms’ Innovation, Constrains and Productivity: the Case of Peru”. Mario Tello.
Octubre, 2014.
No. 381
“The Great Recession: on the Ineffectiveness of Domestic Adjustment Policies
and the Need of Multilateral Arrangements”. Jorge Rojas. Octubre, 2014.
No. 380
“The Political Economy of Growth, Inequality, the Size and Composition of
Government Spending”. Klaus Schimdt-Hebbel y Jose Carlos Tello. Setiembre,
2014.
No. 379
“Sistema educativo peruano: balance y agenda pendiente”. Guillermo Jopen,
Walter Gómez y Herbert Olivera. Setiembre, 2014.
Departamento de Economía - Pontificia Universidad Católica del Perú
Av. Universitaria 1801, Lima 32 – Perú.
Telf. 626-2000 anexos 4950 - 4951
http://www.pucp.edu.pe/economia